Frequently Asked Questions
March 2007
By Althea Zanecosky, MS, RD, LDN
Mid-Atlantic Dairy Association
Q: A good amount of research done on the nutritional benefits of dairy seems to be industry-sponsored. Aren’t these studies product biased?
A: Recently a few Boston-based researchers reported that industry-sponsored nutritional beverage studies were more likely to report “favorable results” for the sponsors than those with no industry support and concluded the reports were biased.
Reviewers from the Children's Hospital of Boston and the Center for Science in the Public Interest (CSPI) started with 538 scientific articles from medical journals that evaluated the health benefits of soft drinks, juice and milk. The studies were published from 1999 to 2003. Using their specific inclusion/exclusion criteria, the 538 studies were then narrowed down to 206. Of these 206 studies, only 111 declared financial sponsorship: 22 percent were funded entirely by industry, 47 percent had no industry funding, and 32 percent had mixed funding. The reviewers then calculated that the 22 percent declaring an industry-only source of funding were four to eight times more likely to report positive conclusions for the sponsors than studies with no industry funding. The report was published on-line in the journal Public Library of Science – Medicine.
It’s important to keep in mind the following:
- The studies used in the analysis were all published in peer-review journals, meaning independent and anonymous experts had already passed their expert eyes over the studies. The studies were published with supporting references and fit in with the results of other similar published research.
- When industry plans to fund a study, it is natural that it would select a product with potentially favorable aspects. National Dairy Council, the scientific branch of the dairy industry, aims for studies that will stress the benefits of dairy products.
- When funders like the American Beverage Association and National Dairy council support research, decisions about the design, statistical analysis, results or conclusions are left entirely to the investigators to ensure that no bias exists.
- When there are no results or negative results, these studies may go unpublished for a variety of reasons. The company that sponsored the research may go back to the drawing board as there may be no need to draw attention to something that doesn't work. Or, while of no use to the company, the study may alert competitors to sensitive areas of research.
- Research is driven by industry. Sources of independent funding for university studies are small and many are very grateful for industry sponsorship. Such partnerships are vital, and companies that put money up for university-based research should be commended.
- In an accompanying perspective article in the same journal, Martijn Katan, professor of nutrition at Vrije Universiteit Amsterdam in the Netherlands, noted that a "blanket condemnation of industry-supported research" isn't the answer. He said partnerships with industry allowed him "to discover things that I could not have found otherwise." Katan and his colleagues discovered the effects of trans fatty acids on heart-disease risk thanks to the expertise of Unilever, and the cholesterol-raising factor in unfiltered coffee with backing from Nestle.
What’s important to conclude is that all research studies should be viewed as one part of the puzzle about food and health and not final answers to questions. Conclusions about any food must come from many studies - not one.
For more information about dairy research go to National Dairy Council:
http://www.nationaldairycouncil.org/NationalDairyCouncil/
Martinj Katan’s remarks can be found at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1761050